java - which is the best way to read data from a table in a pdf?

Question

I want to read the data from the table in this PDF.

PDF

I had thought about reading the PDF, exporting it to an Excel and then use the data. The problem of reading the pdf and exporting it to Excel is that there are elements of columns that move to empty columns because I read with Apache Poi, and in this way the whole PDF is saved in a string.

Another way was to read exact coordinates data, but I do not think it's a very good option.

Could someone advise me? Which way is better or some new way?

Answer 1

I've had the best luck using Xpdf pdftotext with a combination of the -layout and -table options.

Xpdf Link

You would call like this:

pdftotext -table c:	empENaB20180317.pdf c:	empoutput.txt

You could then parse by getting the starting column position from the header on each page.

Antoher good option is PDFBox it may extract the text in a format you can use without having to call a separate command line app.